基于种群协同进化算法的固定极性动态逻辑电路功耗优化

基于电路的动态逻辑实现形式,建立了固定极性XNOR/OR电路低功耗极性优化问题的数学模型;针对传统遗传算法(TGA)和量子算法(TQA)的优势和不足,借鉴合作型协同进化思想,提出了种群协同进化算法(PCEA)。该算法包含主体种群和小规模的量子比特种群,采取两种群并行进化、统一评估和主体种群择优重组的进化策略。主体种群采用包括选择、交叉和变异在内的常规进化方式。量子比特种群采用均匀进化和多次测量的进化方式,以便得到一组尽可能均匀覆盖解空间的个体补充到主体种群,避免算法出现“早熟”现象。最后,8个MCNC Benchmark 电路的测试结果表明了PCEA的优化效果及其稳定性。     

最小二乘支持向量机的参数优化及其应用

针对最小二乘支持向量机的多参数带来的参数寻优问题,将进化算法(遗传算法和PSO算法)应用其中,通过Sinc函数的测试,成功地实现了多参数的联合优化;将这一方法应用到德士古炉温软测量建模中,采用来自工业现场的实测数据进行仿真,将两种方法的仿真结果与常用的BP神经网络进行比较,可以看出两种算法都较好地解决了最小二乘支持向量机的参数优化问题.     

基于灾变型文化算法的不确定条件下中间存储时间有限Flow Shop调度

针对某些不稳定的中间产品,考虑其在中间储罐存储时间有限的调度问题。用粗糙变量表示不确定的加工时间,在粗糙规划 理论的基础上建立相应的粗糙调度模型,并将其清晰等价化。设计了具有针对性的信仰空间及其与种群空间进行沟通的渠道,并在 此基础上仿照自然界的演化规则为解决进化停滞的问题引入了灾变算子,提出了一种新的灾变型文化算法。仿真结果表明了本文所 提出模型和算法的可行性及有效性。与改进遗传算法和无灾变算子的文化算法的对比研究表明了灾变型文化算法具有更好的求解性 能。     

约束差分免疫克隆算法及其在汽油调合优化中的应用

针对约束多目标优化问题,提出了一种新型的约束多目标优化算法。该算法采用了一种新型约束处理方式,先通过约束违反门限截取种群再依据约束与目标函数值针对不同情况实现对个体的优劣划分。本算法将差分进化与免疫克隆机制相融合,既利用了差分进化从全局角度进行搜索的特点,又利用了免疫克隆机制从优秀个体出发进行局部再寻优搜索的优点,扩大了算法搜索的广度与深度。测试结果表明该算法相比快速非支配排序遗传算法（NSGA II）具有非常优秀的收敛性与分布性。将提出的算法应用于实际的汽油调合优化中,进一步验证了算法的有效性,可有效减少成本,提高产品质量。     

投影遗传算法

针对一般遗传算法在求解优化问题的多个优化解时的局限性,结合曲面投影与遗传算法的基本思想,提出了一种基于曲面投影的投影遗传算法用于解决此类问题,数值验证的结果表明了本算法的有效性和实用性。     

基于开放式遗传算法的图像阈值选取

引入了开放式遗传算法的理论，使种群在一个开放的环境中进化，增加了种群的多样性。同时对交叉、变异操作进行了改进，避免了进化过程中种群的退化现象，从而有效克服了遗传算法的早熟问题又提高了遗传算法的收敛性能。文章以最小误差法为例，对比了本文算法和简单遗传算法在阈值处理中的性能，并用实验证明了本文算法的可行性。     

基于遗传算法的机场调度优化算法

随着航班数量的不断增长,航空管理系统已不堪重负,机场容量将成为航空运输发展的瓶颈.为了解决机场容量不足问题,本文将机场调度问题分为杌位分配和滑行道分配两个过程,设计了适合于求解机位分配和滑行道分配问题的遗传算法.对停机位分配问题,在遗传进化过程中为促进算法收敛,采用贪婪算法对种群进行优化,并引入模拟退火思想对适应度函数进行修正.对滑行道分配问题,为适合遗传算法求解,首先将问题转化为图的形式,并设计了相应的遗传编码方式.数值模拟实验表明所提算法能够比较有效地解决机位分配和滑行道分配问题.     

基于小波变换的多分辨率CT/MRI图像配准

本文以仿射模型为基础将区域配准问题转化成优化问题,然后给出小波金字塔分解用于配准的一般框架,并与常规金字塔分解法进行了分析比较;针对配准过程的多变量非线性优化问题,结合混沌理论和遗传算法提出了改进的全局优化的混沌遗传算法.这种算法将混沌序列嵌入标准遗传算法,对标准遗传算法的个体进行混沌化处理,利用混沌的遍历特性来改进遗传算法的鲁棒性.     

基于自适应差分进化算法的间歇反应动态优化求解

为了求解间歇反应动态优化问题,提出了一种自适应差分进化算法(Self-Adaptive Differential Evolution, SADE)。在SADE算法中,每个个体都拥有自己的控制参数。该算法在对原优化问题进行差分进化搜优的同时,以权重大小来评价各个控制参数的优劣,并以加权控制参数作为控制参数的进化方向,实现其自适应调整。结果表明SADE算法收敛速度快、求解精度高。将SADE算法应用于两个典型的间歇反应动态优化问题中,取得了较好的优化效果;同时,分析了时间离散度对优化结果的影响。     

基于混合编码遗传算法的最优跟踪门

跟踪算法优化可以提高跟踪质量,选择恰当的跟踪门是优化跟踪算法的关键措施之一。本文提出了一种基于混合编码的遗传算法,用于杂波环境下目标跟踪过程中跟踪门参数的离线优化。该算法将二进制编码与浮点数编码结合,对跟踪门的形状和大小进行混合编码,选择跟踪精度性能指标构造遗传算法的适应度函数,以此将跟踪算法的优化问题转化为遗传算法寻优,在不同杂波环境下优化跟踪门参数设置。     

The diagnostic rules of peripheral lung cancer preliminary study based on data mining technique

Objective： To discuss the clinical and imaging diagnostic rules of peripheral lung cancer by data mining technique, and to explore new ideas in the diagnosis of peripheral lung cancer, and to obtain early-stage technology and knowledge support of computer-aided detecting （CAD）. Methods： 58 cases of peripheral lung cancer confirmed by clinical pathology were collected. The data were imported into the database after the standardization of the clinical and CT findings attributes were identified. The data was studied comparatively based on Association Rules （AR） of the knowledge discovery process and the Rough Set （RS） reduction algorithm and Genetic Algorithm（GA） of the generic data analysis tool （ROSETTA）, respectively. Results： The genetic classification algorithm of ROSETTA generates 5 000 or so diagnosis rules. The RS reduction algorithm of Johnson＇s Algorithm generates 51 diagnosis rules and the AR algorithm generates 123 diagnosis rules. Three data mining methods basically consider gender, age, cough, location, lobulation sign, shape, ground-glass density attributes as the main basis for the diagnosis of peripheral lung cancer. Conclusion： These diagnosis rules for peripheral lung cancer with three data mining technology is same as clinical diagnostic rules, and these rules also can be used to build the knowledge base of expert system. This study demonstrated the potential values of data mining technology in clinical imaging diagnosis and differential diagnosis.     

自适应杂交NGA及其在多值电路分析中的应用

提出了一种具有自适应交叉算子的ＮＧＡ算法（简记为ＡＣＮＧＡ），以解决具有多模态函数的参数优化问题，ＡＣＮＧＡ能随着解空间参数结构的变化，自适应地选择不同的杂交算子并确定相应的杂交率，使解群在保持多样性的前提下，以较收敛率得到最优解集。该算法可求解非线性方程组，并主解具有任意约束关系的非线性电阻电路的所有解。     

基于多代竞争遗传算法的数值函数优化

目的 对标准遗传算法的过早收敛问题进行改进。方法保持种群的多样性,将上几代个体中的一部分与本代共同参与竞争,提出多代竞争遗传算法,通过理论分析和对数值函数优化证明该算法的有效性。结果推导出多代竞争遗传算法的模式定理,经验证明显优于标准遗传算法。结论多代竞争遗传算法有利于保持种群的多样性,避免了过早收敛。     

3D Image Analysis and Artificial Intelligence for Bone Disease Classification

In order to prevent bone fractures due to disease and ageing of the population, and to detect problems while still in their early stages, 3D bone micro architecture needs to be investigated and characterized. Here, we have developed various image processing and simulation techniques to investigate bone micro architecture and its mechanical stiffness. We have evaluated morphological, topological and mechanical bone features using artificial intelligence methods. A clinical study is carried out on two populations of arthritic and osteoporotic bone samples. The performances of Adaptive Neuro Fuzzy Inference System (ANFIS), Support Vector Machines (SVM) and Genetic Algorithm (GA) in classifying the different samples have been compared. Results show that the best separation success (100 %) is achieved with Genetic Algorithm.     

Diabetic foot wound care algorithms

The purpose of this article is to present reference guidelines to assist clinicians when treating diabetic patients with foot wounds. Diabetic patients with limb-threatening foot ulcers often have multiple coexisting medical conditions that frequently become impediments to the resolution of foot wounds. Each foot wound is unique and its etiology is multifactorial; therefore, each foot wound should be managed differently. The treatment algorithm presented in this article is divided into three categories: Algorithm I describes the treatment of septic foot wounds, which may be considered true podiatric surgical emergencies; Algorithm II describes the treatment of ischemic foot ulcers or gangrene with or without underlying osteomyelitis; and Algorithm III describes the treatment of neuropathic foot ulcers with or without underlying osteomyelitis.     

A NEW METHOD OF IMAGE PROCESSING TECHNIQUE APPLIED TO CHROMOSOME RECOGNITION AND COUNTING

This paper describes a new method based on Four Neighbor Distance Transformation (FNDT) and Equal Diagonal Algorithm (EDA) to extract the medial axis and locate the centromere of the chromosome Compared to the Classical Thinning Algorithm and Four Neighbor Distance Transformation, the new method (FNDT-DEA) is more noise-tolerant, simpler in programming and faster in execution. The FDNTEDA provides the connective and one-pixel thick medial axis representing the length of chromosome. The Crofton directive parameters were used in the algorithm to locate the centromere in some chromosome. The number of chromosomes in a given cell is obtained with calculating either the Euler Number or the clase-curve's subset. An intergroups classifier has also been designed. The FNDT-DEA gave results quite similar to those given with human assessment. It was concluded that this new method, FNDT-EDA, is appropriate for microcomputer-based analyzing the human chromosomes.     

药动/药效相关性进展及其在中药研究中的应用

近年来,药动/药效(PK-PD)的联合模型在药物研究中越来越重要,除了一些传统的模型外,一些新的方法也被应用于药动/药效联合模型中,在数据统计方面也出现了一些新分析方法,如神经网络分析方法、规划求解与遗传算法相结合法等。本文综述了药动/药效的相关性以及数据分析方法的研究进展及其在中药研究中的展望。     

An Efficient Automated Algorithm to Detect Ocular Surface Temperature on Sequence of Thermograms Using Snake and Target Tracing Function

Functional infrared (IR) imaging is widely adopted in medical field nowadays, with more emphasis on breast cancer and ocular abnormalities. In this article, an algorithm is presented to accurately locate the eye and cornea in ocular thermographic sequences, which were recorded utilizing functional infrared imaging. The localization is achieved by snake algorithm coupled with a newly proposed target tracing function. The target tracing function enables automated localization, allows the absence of any manual assistance before the algorithm runs. Genetic algorithm is used to perform the search for global minimum on the function to produce desired localization. On all the cases we have studied, in average the region encircled by the algorithm covers 92% of the true ocular region. As for the non-ocular region covered, it only accounts for less than 5% of the encircled region.     

Node Status Algorithm for Load Balancing in Distributed Service Architectures at Paperless Medical Institutions

Service architectures are necessary for providing value-added services in telecommunications networks, including those in medical institutions. Separation of service logic and control from the actual call switching is the main idea of these service architectures, examples include Intelligent Network (IN), Telecommunications Information Network Architectures (TINA), and Open Service Access (OSA). In the Distributed Service Architectures (DSA), instances of the same object type can be placed on different physical nodes. Hence, the network performance can be enhanced by introducing load balancing algorithms to efficiently distribute the traffic between object instances, such that the overall throughput and network performance can be optimised. In this paper, we propose a new load balancing algorithm called “Node Status Algorithm” for DSA infrastructure applicable to electronic-based medical institutions. The simulation results illustrate that this proposed algorithm is able to outperform the benchmark load balancing algorithms—Random Algorithm and Shortest Queue Algorithm, especially under medium and heavily loaded network conditions, which are typical of the increasing bandwidth utilization and processing requirements at paperless hospitals and in the telemedicine environment.